Tokenization and text representation · MLD-1110
Sequence packing
Combining multiple examples into longer training sequences to reduce wasted space.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Pack short conversations into a full token block.
Original caveat
Boundaries and attention leakage between examples need explicit handling.
Context and aliases
Level: Intermediate. Stages: Data · Training · Inference.
Tags: NLP · SLM · LLM.
Aliases: Example packing.
Related concepts
Original references
- Hugging Face: chat templates ↗ · Reference link
- Transformers glossary ↗ · Consulted
- Hugging Face: tokenization algorithms ↗ · Consulted
- Original reference URL ↗